We demonstrate how social density-based clustering of WLAN traces can be utilised to detect granular social groups of mobile users within a university campus. Furthermore, the ability to detect such social groups, which can be linked to the learning activities taking place at target locations, provides an invaluable opportunity to understand the presence and movement of people within such an environment. For example, the proposed density-based clustering procedure, which we call Social-DBSCAN, has real potential to support human mobility studies such as the optimisation of space usage strategies. It can automatically detect the academic term period, the classes, and the attendance data. From a large Eduroam log of an academic site, we chose as a proof concept, selected locations with known capacity for the evaluation of our proposed method, which we successfully utilise to detect the regular learning activities at those locations, and to provide accurate estimates about the attendance levels over the academic term period.
展开▼
